Question Speculation: RDNA2 + CDNA Architectures thread

[uzzi38]

All die sizes are within 5mm^2. The poster here has been right on some things in the past afaik, and to his credit was the first to saying 505mm^2 for Navi21, which other people have backed up. Even still though, take the following with a pich of salt.

Navi21 - 505mm^2

Navi22 - 340mm^2

Navi23 - 240mm^2

Source is the following post: https://www.ptt.cc/bbs/PC_Shopping/M.1588075782.A.C1E.html

 

[coercitiv]

I don't see a problem with posting videos from the youtubz. This is the speculation thread after all, not the 100% accurate thread.

This is the speculation thread indeed, not the "invent leaks" thread. That youtuber is making claims about AMD that can easily be invalidated.

Their opinion is no more better or worse then ours, they are just people after all and we'll see where the chips lay after release.

It's one thing to speculate based on opinions and another to present opinions as facts. Look at the slides again, everything in red he claims is almost certain, based on verified info. In fact, here's how he described the slides:

Things that are red are things that I'm almost entirely sure about about, and things in white are things I'm almost sure about. I didn't put anything in this presentation that's a complete guess.

So no opinions, no speculation, just hard "red & white facts" based on leak info.

 

[raghu78]

This is the speculation thread indeed, not the "invent leaks" thread. That youtuber is making claims about AMD that can easily be invalidated.

It's one thing to speculate based on opinions and another to present opinions as facts. Look at the slides again, everything in red he claims is almost certain, based on verified info. In fact, here's how he described the slides:

So no opinions, no speculation, just hard "red & white facts" based on leak info.

Well said. These so called tech youtubers lack any real skill, a real job and do not provide any value to anybody. They do provide lots of entertainment with their funny rubbish. They are out to make easy money and lots of people fall for their tricks.

 

[DJinPrime]

Well said. These so called tech youtubers lack any real skill, a real job and do not provide any value to anybody. They do provide lots of entertainment with their funny rubbish. They are out to make easy money and lots of people fall for their tricks.

Even the "good" youtubers, I don't get the feeling that they have any real education/background in engineering. I miss old anandtech, the reviews and deep dives were really good and written by people who actually know what they're talking about.

 

[DiogoDX]

MLID is trash like coretecks.

EDIT> For leak hardware specs precision.

 

[Stuka87]

Even the "good" youtubers, I don't get the feeling that they have any real education/background in engineering. I miss old anandtech, the reviews and deep dives were really good and written by people who actually know what they're talking about.

Both Ryan and Ian very much know what they are talking about. Yes there has been fewer articles, but the ones we do get still have tons of details in them.

 

[blckgrffn]

Both Ryan and Ian very much know what they are talking about. Yes there has been fewer articles, but the ones we do get still have tons of details in them.

For sure, but 15-20 years ago it seemed like the pace was much higher... of course it seemed like more was happening too. The last five years have been relative snoozers.

 

[amrnuke]

For sure, but 15-20 years ago it seemed like the pace was much higher... of course it seemed like more was happening too. The last five years have been relative snoozers.

Ian and Ryan are great, but I agree, there's just not as much to talk about, as things have moved from revolutionary to evolutionary. Still interesting, but much like automobiles, you're going to either have to appreciate the nuances of the changes from year to year, or find another hobby. CPU and GPU are mature techs and expecting revolution is probably unreasonable.

 

[maddie]

For sure, but 15-20 years ago it seemed like the pace was much higher... of course it seemed like more was happening too. The last five years have been relative snoozers.

It was much higher. We seem to be returning to those days. CPU began a few years ago and now the GPU segment.

 

[kurosaki]

I think we have the chance to relive the good old days in a couple of years. I personally think ARM is going to be huge, not only in the laptop segment, but also desktop.
If Microsoft get their mind together and work on their software for the Aarch64, partially the OS, but also their own software. Improving compilers etc.

News the other day that they have improved x86-64 emulation makes no one happy. If they take the leap, like Apple and invest some time and efforts developing for ARM, we can see a shift on ARM-based gaming stations, while enterprise machines stays a while longer on x86.

With ARM we have the chance go grow with orders of magnitude in performance again. Enough for me to get a little excited anyways.



Future sig:
Snapdragon XT6700 64 core OC'd @ 5GHz stable | Radeon X 10900Xt stock | 20 TB NVME SSD | 256 GB LPDDR6 ram @ 5600MHz

 

[leoneazzurro]

AMD is hyping gaming performance:

https://www.amd.com/en/events/gaming-2020#.X3Z8AiHjtLg.twitter

Let's hope there will be substsance on these claims

 

[Vixis Rei]

This is the speculation thread indeed, not the "invent leaks" thread. That youtuber is making claims about AMD that can easily be invalidated.


It's one thing to speculate based on opinions and another to present opinions as facts. Look at the slides again, everything in red he claims is almost certain, based on verified info. In fact, here's how he described the slides:


So no opinions, no speculation, just hard "red & white facts" based on leak info.

Context is important and understanding what he meant also is.

If he said the red text is nearly accurate and the white text is mostly accurate. you can't call those facts as he never did. the key line is "i'm almost entirely sure about" and "i'm mostly sure about". That makes them "conclusions he's almost sure about / mostly sure about based on the data he's been given". again HE. that word is important because that means that it's speculation/conclusions. You can't call them facts in your last sentence because they never were such a thing. And even still. Facts can't exist yet because we don't know too much info to put gorilla glue on and the product doesn't exist to the outside world yet. He'd have to be a cheesecake to say that. AND even if he did say that, we all know good and well they aren't facts.

He also spent from 1:35 to 3:40 basically mentioning that. He said; "he expects to keep recieving more information, nothings a complete guess, some sources disagreed on some things, he expects to get more information, he doesn't know every detail, some theories he had got debunked but how he believe he now have enough information (to be sure of what he's saying) and will admit if he's wrong".... It could be considereded speculation based on informed data but i mean he's at least attempting to vet his leaks more (which is important if he have any / more then 1) and i don't mind hearing him out one bit because it does not hurt me to at least listen and form my own conclusions based on what he and you all say. Which at the end of the day is what we should be doing after all.

I'd say that if MLID has good leakers. they don't seem to agree with each other a whole lot. lol

Generally speaking. what we should be doing is checking his and everyone else info in detail and date stamps (or chart) to see if it matches up with anything from anywhere or anyone else. If we have people saying that AMD is sending out lots of fake info to lots of places, knowing what his unique info is / came from /similar to could be more important then just floundering around in the dark especially if this is AMD's tactics going forward.

But i still love me some good thoughtful speculation. Keeps my AIO's a chuggan. I read this whole topic (and the ampere thread) from page 1 to current and don't regret a thing.


Edited out my lats paragraph and comment because i deemed it unnecessary mud. if you still want to read that dribble it's in the attached file.

 

[PhoBoChai]

AMD is hyping gaming performance:

https://www.amd.com/en/events/gaming-2020#.X3Z8AiHjtLg.twitter

Let's hope there will be substsance on these claims

There's no need to hope.

8c unified cache, all low latency on the same interconnect and die. Since games rarely push above 8c, most AAA titles will fit nicely within a single 8c complex.

Fun facts: Ryzen latency within a CCX to any core, is much lower than Intel's ringbus to any core.

They are releasing these Zen 3 CPUs first, means reviewers will see how powerful they are and switch over their test rigs, PCIE gen 4 for all, just in time for the RDNA 2 launch.

 

[coercitiv]

If he said the red text is nearly accurate and the white text is mostly accurate. you can't call those facts as he never did.

No no no, he said "almost entirely sure" and "almost sure". He did not say "mostly", don't try to downplay his claims. Certainty is based on either belief or facts, so I will call those facts unless you want us to go towards faith based speculation.

He also spent from 1:35 to 3:40 basically mentioning that. He said; "he expects to keep recieving more information, nothings a complete guess, some sources disagreed on some things, he expects to get more information, he doesn't know every detail, some theories he had got debunked but how he believe he now have enough information (to be sure of what he's saying) and will admit if he's wrong".... It could be considereded speculation based on informed data but i mean he's at least attempting to vet his leaks more

I don't think we're on the same page here. I don't have a problem with channels and/or twitter accounts that claim to have leaks, that's fair game and they will be judged by their accuracy. We have Twitter accounts with great results, we also had newcomers on YouTube that make a complete mess before Zen2 launched.

What I do have a problem with is making a show out of it and adding conflicting claims such as fake Linux drivers into the mix to make the show more appealing. This is conspiracy level stuff, and has already been debunked on the forums. Moreover, his language made it seem as if AMD had internally tested physical Navi chips with 384bit memory interface and decided to go with 256bit instead. Think about the implications of this claim as well, because he's almost entirely sure about it.

It's a show, drowning us in (conflicting) details and commentary to keep the audience watching and increase credibility. He did the same with Ampere in the Nvidia Forced Scarcity video.

• He claimed Nvidia is trying to increase margins by forcing AIBs to cut down on BOM and also increase pricing over MSRP. Except we now know cards such as the Asus TUF Gaming took no shortcuts when it comes to BOM and will sell at MSRP.
• He claimed FE cards would be loss leaders with the purpose of getting better results in reviews over AIB cards, except AIB cards have proven to be faster and cooler.

Again, I won't fault him for claiming Samsung 8nm yields were poor and then reconsidering them as good. That's understandable for someone working with leaks and rumors. The problem is adding false juicy information on top to increase... conversion. That's the Mars fertilizer.

 

[Kuiva maa]

I have watched MLID first video back when it came out and I never bothered with the channel ever again. Reading about his content from posters in this thread tells me my gut feeling to ignore him was right. As for AMD, I don't ever recall a Ryzen generation being marketed to gamers first and foremost, does anyone else have better recollection?

 

[kurosaki]

I have watched MLID first video back when it came out and I never bothered with the channel ever again. Reading about his content from posters in this thread tells me my gut feeling to ignore him was right. As for AMD, I don't ever recall a Ryzen generation being marketed to gamers first and foremost, does anyone else have better recollection?

Something, something Gamecache(TM)? But else, no. Got a feeling they are catching in on Intel, maybe even passes with several %

Fake leak? https://www.guru3d.com/news-story/amd-ryzen-9-5900x-cpu-z-bench-score-leaks-and-it-is-fast.html

 

[Vixis Rei]

No no no, he said "almost entirely sure" and "almost sure". He did not say "mostly", don't try to downplay his claims. Certainty is based on either belief or facts, so I will call those facts unless you want us to go towards faith based speculation.

he says "mostly sure" at 3:20. unless you heard it from another timestamp but i think that's the only one to exist.

To be clear. i'm definintly not trying to straight up defend him to be clear. I just don't believe in throwing out info completely without looking into it. I know how he is and i know of what to throw out of his info. I know enough about tech to know that i know more then him

Certainty can be based on multiple things and not just belief or facts. From a dictionary it means: "the quality or state of being certain especially on the basis of evidence" among other definitions which can be and not limited to "belief and facts" Calling what he said outright facts is misleading and is why i'm posting.

Unless you mean "situationally percieved facts as one was led to believe through various accounts of others"

I don't think we're on the same page here. I don't have a problem with channels and/or twitter accounts that claim to have leaks, that's fair game and they will be judged by their accuracy. We have Twitter accounts with great results, we also had newcomers on YouTube that make a complete mess before Zen2 launched.

We are on the same page. the majority of that paragraph you quoted is nearly completely comprised of exact or near exact quotes from the video in question.

Won't deny the zen2 mess but it wasn't just new youtube channels that caused it all, but it had one of the biggest incidents. I think some of them learned and have lived that down enough.

What I do have a problem with is making a show out of it and adding conflicting claims such as fake Linux drivers into the mix to make the show more appealing. This is conspiracy level stuff, and has already been debunked on the forums. Moreover, his language made it seem as if AMD had internally tested physical Navi chips with 384bit memory interface and decided to go with 256bit instead. Think about the implications of this claim as well, because he's almost entirely sure about it.

Making a show out of it is fine. You do what you have to do to make a living / get an audience on youtube of all platforms. I can't fault anyone for making their living however they must / want to. but agreed that the conflicting claims are not cool.

I too would like to hear more clarity on what he ment by the 384bit navi chip (what level of testing) I don't subscribe to him or tip him in live streams (or even watch them) so that'll be a difficult task. But i'd surely like to hear what you have to say about that and i promise i won't reply to that point. I'm just interested in what you have to say. Information makes me excited.

It's a show, drowning us in (conflicting) details and commentary to keep the audience watching and increase credibility. He did the same with Ampere in the Nvidia Forced Scarcity video.

I'd like to play devils advocate here if you'd so allow. On second thought it's not the Nvidia thread so nevermind

He claimed Nvidia is trying to increase margins by forcing AIBs to cut down on BOM and also increase pricing over MSRP. Except we now know cards such as the Asus TUF Gaming took no shortcuts when it comes to BOM and will sell at MSRP.

I don't think this point is as accurate as it should be. Is it from the same video you posted above? He says at 3:49 "Most of the bills of materials that AIB's have submitted have margins below 40% for any of the cards coolers that are near the retail price." Which means that it does not stop them from selling at / near / below MSRP. His assumption that they'd only increase pricing isn't entirely wrong since he said "most" at the start but if they're willing to accept lower margins then he was largely on point.

He claimed FE cards would be loss leaders with the purpose of getting better results in reviews over AIB cards, except AIB cards have proven to be faster and cooler.

I'll pick at this. He was wrong that the FE edition would be loss leaders. But also fair to say that having a direct from Nvidia reviewed GPU ensures a certain quality over random AIB versions. However i don't think there has been a time when some AIB version wasn't better then the main AMD / Nvidia version at least in recent years. Not putting words in his mouth but that's worth stating against both what he and you said. Also has to be stated that there will be an AIB version / versions somewhere that's worse then Nvidia's version.

Again, I won't fault him for claiming Samsung 8nm yields were poor and then reconsidering them as good. That's understandable for someone working with leaks and rumors. The problem is adding false juicy information on top to increase... conversion. That's the Mars fertilizer.

I'd agree to that specific point.

I do like that he admitted he was wrong and corrected himself. I don't have a problem with juicy information, but false juicy information is as bad as fruit stripe gum.

Also glad my quotes didn't break....

 

[lixlax]

What's your take on the RTX 3070 delay regarding RDNA2?

1) Nvidia knows RDNA2 is not impressive and they are going to steal all the thunder after AMD presentation?
2) They know 3070 isn't very competitive vs RDNA2 and Nvidia wants to avoid comparisons on AMD slides?
3) They have no idea what the GPUs from AMD are like and how they perform (so they can cut the price if needed)?
4) It has nothing to do with AMD and they're just improving drivers and stock?
5) ??

 

[kurosaki]

What's your take on the RTX 3070 delay regarding RDNA2?

1) Nvidia knows RDNA2 is not impressive and they are going to steal all the thunder after AMD presentation?
2) They know 3070 isn't very competitive vs RDNA2 and Nvidia wants to avoid comparisons on AMD slides?
3) They have no idea what the GPUs from AMD are like and how they perform (so they can cut the price if needed)?
4) It has nothing to do with AMD and they're just improving drivers and stock?
5) ??

3! My first thought screams 3! 😁

 

[deathBOB]

What's your take on the RTX 3070 delay regarding RDNA2?

1) Nvidia knows RDNA2 is not impressive and they are going to steal all the thunder after AMD presentation?
2) They know 3070 isn't very competitive vs RDNA2 and Nvidia wants to avoid comparisons on AMD slides?
3) They have no idea what the GPUs from AMD are like and how they perform (so they can cut the price if needed)?
4) It has nothing to do with AMD and they're just improving drivers and stock?
5) ??

They already listed the price of the 3070.

It's obviously 4, both AMD and Nvidia will sell every new GPU they make between now and the holidays thanks to covid.

 

[blckgrffn]

They already listed the price of the 3070.

It's obviously 4, both AMD and Nvidia will sell every new GPU they make between now and the holidays thanks to covid.

Pretty much this for 2020 sales. I feel like AMD as the relatively weaker brand has more to worry about with a weak launch than nvidia. There are so many who will only consider nvidia solutions regardless of their competitiveness. They need to simply see enough value against the previous generation/two generations in benchmarks and you can leave AMD right off the graphs.

If AMD really delivers a solid right hook AND a body blow the only way it will really matter is if they can just flood the shelves with stock (ideally from the top to the bottom of the stack) and nvidia runs out. I would imagine that most people buy and sit on their GPUs for 1-3 years at a time and are most likely to stick with the brand they previously purchased. That's just "typical" customer behavior.

That could be something that actually changes the pie chart.

 

[Heartbreaker]

It's a show, drowning us in (conflicting) details and commentary to keep the audience watching and increase credibility. He did the same with Ampere in the Nvidia Forced Scarcity video.

• He claimed Nvidia is trying to increase margins by forcing AIBs to cut down on BOM and also increase pricing over MSRP. Except we now know cards such as the Asus TUF Gaming took no shortcuts when it comes to BOM and will sell at MSRP.
• He claimed FE cards would be loss leaders with the purpose of getting better results in reviews over AIB cards, except AIB cards have proven to be faster and cooler.

Again, I won't fault him for claiming Samsung 8nm yields were poor and then reconsidering them as good. That's understandable for someone working with leaks and rumors. The problem is adding false juicy information on top to increase... conversion. That's the Mars fertilizer.

MLID is nonsense to a ridiculous level, and he keeps getting proved wrong, and yet a bunch of people keeping treating it like something reliable.

I remember people getting proactively angered with NVidia, because MLID said that DLSS 3.0 in Ampere would just work with any game that has TAA and that NVidia was considering defaulting it to on, so they would score higher in benchmarks.

So people watched the video, then got mad because NVidia was going to cheat by turning on DLSS 3.0 by default.

This was all such obvious nonsense, from nearly every angle. He also claimed Ampere RT performance was 4 Times higher than Turing, possibly in the same video. Some people were angry about this to, since NVidia was wasting too much effort on RT performance.

So Ampere comes out, all the stuff is shown to be total nonsense, and people just go back to lapping up his new nonsense without a thought, about how inaccurate he has been in the past.

 

[blckgrffn]

MLID is nonsense to a ridiculous level, and he keeps getting proved wrong, and yet a bunch of people keeping treating it like something reliable.

I remember people getting proactively angered with NVidia, because MLID said that DLSS 3.0 in Ampere would just work with any game that has TAA and that NVidia was considering defaulting it to on, so they would score higher in benchmarks.

So people watched the video, then got mad because NVidia was going to cheat by turning on DLSS 3.0 by default.

This was all such obvious nonsense, from nearly every angle. He also claimed Ampere RT performance was 4 Times higher than Turing, possibly in the same video. Some people were angry about this to, since NVidia was wasting too much effort on RT performance.

So Ampere comes out, all the stuff is shown to be total nonsense, and people just go back to lapping up his new nonsense without a thought, about how inaccurate he has been in the past.

His bank account thanks them all. Seriously, this has to be way easier than many of the other jobs available.

 

[Vixis Rei]

What's your take on the RTX 3070 delay regarding RDNA2?

1) Nvidia knows RDNA2 is not impressive and they are going to steal all the thunder after AMD presentation?
2) They know 3070 isn't very competitive vs RDNA2 and Nvidia wants to avoid comparisons on AMD slides?
3) They have no idea what the GPUs from AMD are like and how they perform (so they can cut the price if needed)?
4) It has nothing to do with AMD and they're just improving drivers and stock?
5) ??

I feel like it's probably a few things at once.
They probably feel like since RDNA2 launches so late they can probably sell more 3080's / 3090's before AMD meteor showers on their parade.

They may have already listed the price of the 3070 but releasing after the AMD showcase allows them to adjust prices if needed and not be so caught out of position. stamping any silly amd tricks.

Could also be building up stock of 3070's to have so the launch doesn't seem so anemic for a potentially higher volume selling card. cause i don't doubt that AMD and Nvidia will see through all of them.

So maybe mostly #3? maybe.... I dunno if i was an evil cooperation i'd be thinking all this...

 

[DisEnchantment]

[PATCH 1/7] drm/amdgpu: add Green_Sardine APU flag

[PATCH 4/7] drm/amdgpu: add gfx support for green_sardine (v2)

[PATCH 4/7] drm/amdgpu: add gfx support for green_sardine (v2)

Oh hello ... Green Sardine APU

It is Vega derived APU like Renoir.

 

[Vixis Rei]

[PATCH 1/7] drm/amdgpu: add Green_Sardine APU flag

[PATCH 4/7] drm/amdgpu: add gfx support for green_sardine (v2)

[PATCH 4/7] drm/amdgpu: add gfx support for green_sardine (v2)

Oh hello ... Green Sardine APU

It is Vega derived APU like Renoir.

I would absolutely love it if they put these codenames, number hashes and a picture related to the codenames on the bottom of the retail box. Would be hilarious and i'd be proud to own said box!